deep-research

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Deep Research

深度研究

Produce thorough, cited research reports from multiple web sources using firecrawl and exa MCP tools.

借助firecrawl和exa MCP工具，基于多个网页来源生成详尽的带引用研究报告。

When to Activate

激活场景

User asks to research any topic in depth
Competitive analysis, technology evaluation, or market sizing
Due diligence on companies, investors, or technologies
Any question requiring synthesis from multiple sources
User says "research", "deep dive", "investigate", or "what's the current state of"

用户要求对任意主题进行深度研究
竞品分析、技术评估或市场规模测算
对企业、投资者或技术开展尽职调查
任何需要整合多来源信息的问题
用户提及“研究”“深度剖析”“调查”或“当前现状如何”

MCP Requirements

MCP 要求

At least one of:

firecrawl —

firecrawl_search

firecrawl_scrape

firecrawl_crawl

exa —

web_search_exa

web_search_advanced_exa

crawling_exa

Both together give the best coverage. Configure in

~/.claude.json

~/.codex/config.toml

至少满足以下其中一项：

firecrawl —

firecrawl_search

firecrawl_scrape

firecrawl_crawl

exa —

web_search_exa

web_search_advanced_exa

crawling_exa

同时使用两者可获得最佳覆盖范围。可在

~/.claude.json

或

~/.codex/config.toml

中进行配置。

Workflow

工作流程

Step 1: Understand the Goal

步骤1：明确目标

Ask 1-2 quick clarifying questions:

"What's your goal — learning, making a decision, or writing something?"
"Any specific angle or depth you want?"

If the user says "just research it" — skip ahead with reasonable defaults.

提出1-2个快速澄清问题：

“你的目标是什么——学习知识、制定决策还是撰写内容？”
“是否有特定的研究角度或深度要求？”

如果用户表示“直接研究即可”——则采用合理默认设置继续。

Step 2: Plan the Research

步骤2：规划研究内容

Break the topic into 3-5 research sub-questions. Example:

Topic: "Impact of AI on healthcare"
- What are the main AI applications in healthcare today?
- What clinical outcomes have been measured?
- What are the regulatory challenges?
- What companies are leading this space?
- What's the market size and growth trajectory?

将主题拆解为3-5个研究子问题。示例：

主题：“AI对医疗保健的影响”
- 当前医疗保健领域的主要AI应用有哪些？
- 已测得哪些临床效果？
- 面临哪些监管挑战？
- 该领域的领先企业有哪些？
- 市场规模及增长趋势如何？

Step 3: Execute Multi-Source Search

步骤3：执行多源搜索

For EACH sub-question, search using available MCP tools:

With firecrawl:

firecrawl_search(query: "<sub-question keywords>", limit: 8)

With exa:

web_search_exa(query: "<sub-question keywords>", numResults: 8)
web_search_advanced_exa(query: "<keywords>", numResults: 5, startPublishedDate: "2025-01-01")

Search strategy:

Use 2-3 different keyword variations per sub-question
Mix general and news-focused queries
Aim for 15-30 unique sources total
Prioritize: academic, official, reputable news > blogs > forums

针对每个子问题，使用可用的MCP工具进行搜索：

使用firecrawl时：

firecrawl_search(query: "<子问题关键词>", limit: 8)

使用exa时：

web_search_exa(query: "<子问题关键词>", numResults: 8)
web_search_advanced_exa(query: "<关键词>", numResults: 5, startPublishedDate: "2025-01-01")

搜索策略：

每个子问题使用2-3种不同的关键词变体
混合使用通用查询和新闻类查询
目标获取15-30个独特来源
优先级：学术资源、官方信息、权威新闻 > 博客 > 论坛

Step 4: Deep-Read Key Sources

步骤4：深度阅读关键来源

For the most promising URLs, fetch full content:

With firecrawl:

firecrawl_scrape(url: "<url>")

With exa:

crawling_exa(url: "<url>", tokensNum: 5000)

Read 3-5 key sources in full for depth. Do not rely only on search snippets.

针对最具价值的URL，获取完整内容：

使用firecrawl时：

firecrawl_scrape(url: "<url>")

使用exa时：

crawling_exa(url: "<url>", tokensNum: 5000)

深度阅读3-5个关键来源，不要仅依赖搜索摘要。

Step 5: Synthesize and Write Report

步骤5：整合并撰写报告

Structure the report:

markdown

undefined

报告结构如下：

markdown

undefined

[Topic]: Research Report

[主题]：研究报告

Generated: [date] | Sources: [N] | Confidence: [High/Medium/Low]

生成日期：[日期] | 来源数量：[N] | 可信度：[高/中/低]

Executive Summary

执行摘要

[3-5 sentence overview of key findings]

[3-5句话的核心发现概述]

1. [First Major Theme]

1. [第一大主题]

[Findings with inline citations]

Key point (Source Name)
Supporting data (Source Name)

[带有内联引用的研究结果]

核心观点 (来源名称)
支撑数据 (来源名称)

2. [Second Major Theme]

2. [第二大主题]

...

3. [Third Major Theme]

3. [第三大主题]

...

Key Takeaways

关键结论

[Actionable insight 1]
[Actionable insight 2]
[Actionable insight 3]

[可落地见解1]
[可落地见解2]
[可落地见解3]

Sources

参考来源

Title — [one-line summary]
...

标题 — [一句话摘要]
...

Methodology

研究方法

Searched [N] queries across web and news. Analyzed [M] sources. Sub-questions investigated: [list]

undefined

在网页和新闻平台上执行了[N]次查询，分析了[M]个来源。研究的子问题包括：[列表]

undefined

Step 6: Deliver

步骤6：交付成果

Short topics: Post the full report in chat
Long reports: Post the executive summary + key takeaways, save full report to a file

短主题：在聊天中发布完整报告
长报告：发布执行摘要+关键结论，将完整报告保存为文件

Parallel Research with Subagents

借助子代理进行并行研究

For broad topics, use Claude Code's Task tool to parallelize:

Launch 3 research agents in parallel:
1. Agent 1: Research sub-questions 1-2
2. Agent 2: Research sub-questions 3-4
3. Agent 3: Research sub-question 5 + cross-cutting themes

Each agent searches, reads sources, and returns findings. The main session synthesizes into the final report.

针对宽泛主题，使用Claude Code的Task工具实现并行处理：

并行启动3个研究代理：
1. 代理1：研究子问题1-2
2. 代理2：研究子问题3-4
3. 代理3：研究子问题5 + 跨领域主题

每个代理负责搜索、阅读来源并返回研究结果，主会话将这些结果整合成最终报告。

Quality Rules

质量规则

Every claim needs a source. No unsourced assertions.
Cross-reference. If only one source says it, flag it as unverified.
Recency matters. Prefer sources from the last 12 months.
Acknowledge gaps. If you couldn't find good info on a sub-question, say so.
No hallucination. If you don't know, say "insufficient data found."
Separate fact from inference. Label estimates, projections, and opinions clearly.

所有主张均需标注来源。不得出现无来源的断言。
交叉验证。若仅有单一来源提及某内容，需标记为未验证。
时效性优先。优先选择过去12个月内的来源。
如实说明信息缺口。若无法找到某子问题的优质信息，需明确告知。
禁止虚构内容。若信息未知，需说明“未找到足够数据”。
区分事实与推论。明确标记估算、预测和观点内容。

Examples

示例

"Research the current state of nuclear fusion energy"
"Deep dive into Rust vs Go for backend services in 2026"
"Research the best strategies for bootstrapping a SaaS business"
"What's happening with the US housing market right now?"
"Investigate the competitive landscape for AI code editors"

“研究核聚变能源的当前现状”
“深度剖析2026年Rust与Go在后端服务中的对比”
“研究启动SaaS业务的最佳策略”
“美国房地产市场当前状况如何？”
“调查AI代码编辑器的竞争格局”